The U.S. National Transportation Safety Board is defending itself against accusations it skewed a study that preceded the shutdown of 26 so-called Chinatown bus operations in the Northeast.
Anomalies in how the board classified motorcoach operators and calculated fatality rates raise doubts about its conclusion that curbside companies, including most Chinatown carriers, are about seven times more probable to have passenger deaths than companies using terminals, according to the Reason Foundation, a Los Angeles-based advocacy group for limited government.
Safety board officials in interviews confirmed they classified Greyhound, the largest U.S. terminal-based operator, as a curbside company. Counting Greyhound as a terminal operator would narrow the gap in fatality rates, according to data compiled by Bloomberg.
“They committed statistical malpractice by refusing to make the data available to the public and by presenting the ‘seven times’ figure without the gigantic qualification” that the margin of error could have eliminated the difference between the two carrier types, said Aaron Brown, a risk manager trained in statistical mathematics.
Brown, who works for AQR Capital Management in Greenwich, Connecticut, first critiqued the NTSB study last July in an article on the financial news website Minyanville.
The recalculations don’t alter the finding that curbside operators have been more dangerous, Eric Weiss, a board spokesman, said.
The NTSB study was requested by Senator Charles Schumer and Representative Nydia Velazquez, both New York Democrats, two days after a bus traveling to New York’s Chinatown crashed in the Bronx in March 2011, killing 15 people. The NTSB, better known for crash investigations that can take more than a year, produced the bus report in a little more than seven months.
NTSB Chairman Deborah Hersman released the findings at a news conference in Manhattan’s Chinatown on Oct. 31, 2011, surrounded by buses loading passengers waiting on sidewalks. Seven months after that, the U.S. Transportation Department conducted its largest one-day safety sweep on bus companies, closing 26 Chinatown lines in a single day.
The Federal Motor Carrier Safety Administration’s action against the Chinatown carriers was based on its own yearlong probe that predated the NTSB’s findings, said Duane DeBruyne, an agency spokesman.
The NTSB counted as curbside carriers FirstGroup Plc’s Greyhound Lines Inc., the largest U.S. intercity bus company, as well as Peter Pan Bus Lines Inc., Martz Trailways, Adirondack Trailways, Fullington Trailways and Greyhound’s Americanos subsidiary.
All of those appear to operate as terminal-based companies, said Jim Epstein, a television producer with Reason who wrote about the study in a Reason.com article posted May 7. He said he called each company before writing his critique.
“Statistical studies have to be rigorous when it comes to using data,” Epstein said. “The fact that the NTSB misclassified so many bus companies shows something went wrong.”
Epstein, in his article, said Greyhound accounted for 24 of 37 of the fatal accidents attributed to curbside operators in the six years studied and 34 of the 51 fatalities.
He also questioned the NTSB’s decision to calculate fatality rates by dividing deaths by the number of buses each company operates, rather than how many miles they drove. The agency said in its report it couldn’t use the latter method because many bus companies don’t report miles driven to the U.S. Department of Transportation.
The NTSB gave companies with hundreds of buses criss-crossing the country the same weight as companies with a single bus, Brown said.
“They don’t have enough data, they have no data on the length of bus trips, and there just aren’t enough accidents,” Brown said in an interview. “Averaging companies is a very dubious technique.”
Greyhound, with 1,515 buses, dominates the curbside category in the safety board study, while New Jersey Transit, with 2,172 buses, is the largest company counted as a terminal operator.
“They’re really comparing Greyhound to New Jersey Transit” instead of commercial curbside carriers with terminal operators, Brown said.
Timothy Stokes, a Firstgroup Plc spokesman, declined to comment on the NTSB’s research.
Counting the companies cited by Epstein as terminal operators would narrow the ratio of fatal-crash risk between curbside and terminal operators to 4.6 to 1, according to Bloomberg calculations from study data provided by the safety board.
Just moving Greyhound would slightly increase the fatality rate of terminal operators while not changing the rate for curbside carriers. That’s because Greyhound’s fatal-crash rate, 1.39 per 100 buses, was almost exactly the curbside companies’ average during the period evaluated.
The NTSB stands by the study and its conclusions, said Weiss, the agency spokesman. The study focused on the safety record of the companies, not their size, route system or types of buses, he said.
“Safe bus operations should not depend on whether passengers are picked up or dropped off at a traditional bus station or down on the corner,” Weiss said. “It should depend upon safety culture and other issues such as driver training and bus maintenance.”
The agency counted Greyhound as a curbside carrier because of its BoltBus brand, the second-biggest U.S. curbside operator after Stagecoach Group Plc’s Megabus.com.
Because there is no standard government definition of “curbside,” the board’s researchers included any company in the category that operated at least one curbside route.
The agency looked at curbside carriers more broadly, rather than those based in Chinatown, because that’s what the lawmakers asked it to do, Weiss said. NTSB research reports can take six to 18 months to produce, depending on the complexity of the topic and availability of staff resources, he said.
BoltBus, which Greyhound operates as a joint venture with Peter Pan, operated fewer than 100 buses in 2011, Stokes said.
The fatal-crash rate was only one measure used to compare the record of curbside bus companies in the 67-page report, Weiss said. Researches also looked at non-fatal crashes, inspection records and traffic violations, among other metrics, he said.
Three fatal crashes on the U.S. East Coast last year involved Chinatown buses. Two, bound for New York’s Chinatown, were subject of NTSB investigations.
World Wide Travel of Greater New York Ltd., the company involved in the crash that spurred Schumer and Velazquez to ask NTSB for the study, had an accident rate of 5.7 per million miles traveled, 375 percent higher than the threshold the motor carrier regulator deems unacceptable.
The company was rejected by the U.S. Defense Department when it tried to win a contract to carry military personnel in 2009.
Three Chinatown operations in New York and Philadelphia were the primary targets in the May 31, 2012, U.S. safety crackdown, while 26 individual companies were shuttered.
The increased enforcement continued this year, when Fung Wah, the largest Chinatown bus company operating between New York and Boston, was shut after the Massachusetts Department of Public Utilities found cracked bus frames and attempted repairs that appeared to worsen the defects.